Puzzle Game Magic Secrets

Description

In this 2019 GDC session, My Dog Zorro’s Brett Taylor approaches the art and science of puzzle game design through the unstoppably cool lens of human cognition.

Linelight on Steam

Inspirations For This Talk

• Thinking Fast and Slow book

My Notes

What makes a puzzle fun?

• 03:19: What makes a puzzle fun?
  • When the player gets to exhibit mastery, skill, or wit.
    • i.e. when you create a situation that causes the player to feel clever.

What are mechanics?

03:30:

Working Memory

• 03:48: Human RAM: The amount of power that the player has to hold info in their head and use that to solve puzzles.
  • It is finite
• 04:35: Working memory and Flow state:
  • 
  • 
  • Requiring too much or too little working memory produces anxiety and boredom.
  • Flow state is fun.
  • boredom and anxiety are not fun.
• As the player gains experience, they are able to handle more complexity while using less working memory.
  • They are less likely to feel overwhelm.

Handcrafted vs. procedural puzzles

• 08:40:
• 

Noise

• ⭐ 11:11: Noise is anything that takes away from the player’s working memory
• Noise is effectively our dial to turn to make sure that the player’s working memory is at a level that is still fun.
  • Noise makes puzzles artificially harder.
  • If left unchecked noise will pollute your game.
• 15:29: Noise, by definition is not necessary to the solution of the puzzle. It doesn’t have to be in the puzzle.
  • But it does contribute to the experience that the puzzle creates.

Seven lessons learned from making Linelight

1. Simplify

• Simplifying leads to cleaner, tighter puzzles.
• How:
  • Describe the solution.
  • Remove anything you didn’t mention.

2. Cut Pointless Levels

• 20:54
• 21:35: About mediocrity
  • Don’t be afraid of not being good enough.
  • 
• It’s hard to admit a level is pointless. Especially if there is nothing better to replace it.
• Each level should be educational, interesting, fun, or unique.

3. You’re the expert player

23:05

• You’re the expert player (controls wise)
  • Recognize that the puzzles will probably be easier for you than for your players. So you will probably need to make it easier for your players.
• ⭐: ==Knowing the solution and not being able to execute it is not fun!==
• Try this: Play through your levels with your feet. If you can’t, then make your puzzle easier. (specifically puzzle games, not action games)

4. Keep action and puzzles separate

• 24:58
• This is specifically single (or few) solution puzzles.
• Action adds randomnesss and Noise
• Puzzle-solving requires patterns and consistency
• The player needs an idea of what they don’t yet know
• It should be unambiguous if the level requires action or puzzle skills.
• Suggestion: Make action levels optional.

5. Make the solution unambiguous

26:54

• If the solution is ambiguous, it damages the player’s trust in the game.
• When the player solves the puzzle, they should feel like they solved it:
  • They should feel the satisfaction of knowing that they solved it.
  • they should not feel like they solved it by accident or solved it by trial-and-error.
• 29:46: Ambiguity, puzzle-solving and action:
  • It is not a fun experience when the player doesn’t know if something is impossible.
  • It should be unambiguous if the player needs action skills, or puzzle-solving.
    • Puzzle-solving: I can’t do this because I haven’t found the solution to the puzzle yet.
    • Action: I can’t do this now, but I can practice and get better.
    • Not fun: I don’t know if I can do this or not. I don’t know if I need to solve a puzzle. I don’t know if it’s my fault that I’m failing at this.

6. Protect Player Trust

30:51

• Players tend to assume that a puzzle game doesn’t require any reflexes.
• But if you break this rule even once:
  • Now EVERY puzzle maybe requires reflexes

7. Choose Fun Over Exhaustive Design

32:02

• Fun: exhibit wit, skill, mastery 👆🏼
  • fun for the designer ≠ fun for the player
• Exhaustive Design: There is one of every possible puzzle.
  • Sometimes this is incompatible with fun.
  • Braid and The Witness are exhaustive design.

The Mechanics Design the Levels

• 38:03: “Where did you come up with this level?”
  • “I didn’t. The mechanics did.”
• My strategy: Emergent Design
  • 1. Create the mechanics
  • 2. Use the mechanics together.
    • Let the mechanics interact with each other.
  • 3. Now you have interesting puzzles!
• 

How do you create the mechanics?

• 38:40: Experience, intuition, and luck
• Create mechanics that…
  • are diverse
  • could interact in many ways
  • excite you. (are fun)
• Get rid of any mechanics that are not fun.
• Experiment a lot!

  • “The more darts you throw, the more bullseyes you will hit.”

• The goal is to have few mechanics that yield many puzzles.
• It’s a good sign if you are surprised by your mechanics.
  • This means you probably found something that is fun!

Solution Sentences

41:27:

• Each puzzle has a solution sentence which is a sentence that explains the solution to the puzzle. (Without any of the noise).
  • Each solution sentence is like a magic trick’s secret.
• If you use solution sentences then elegance will be inevitable.
• simplicity is intrinsic to elegance.
• ❓: 45:24: How do you create solution sentences?
  • Force mechanics to interact with each other, and see what happens. Emergent Design.
  • Ask “What if ____”
  • 
    • This is a spreadsheet, combining game mechanics together to form solution sentences.

Emergent Design in Puzzle Games

• 47:51: Game designers don’t design puzzles. They design the mechanics, and the mechanics lead to the puzzles.
  • This is Emergent Design in Game Development.
• 
  • Done rightly, it should actually be easier to create a puzzle, than to solve it.

Q and A

• 53:55: How do you know which content needs to be thrown out?
  • It’s only after I’ve created enough content, that I know what should be cut.
• 56:31: Control paradigms: console vs. PC vs. mobile etc.
• 58:34: If you created a procedural puzzle game, how would you approach it?

Transcript

---